HyperMarkdown Caches Code and Tables to Beat Streaming Renderers

HyperMarkdown renderer achieves 1.8x to 11x speedup over competitors by caching settled code lines and table rows during AI streaming.

TruthFoundry News Desk

Stands on 5 placed sources from 2 publishers.

HyperMarkdown is 1.8 to 11 times faster than the nearest streaming renderer across the benchmark suite, with performance gains varying by workload type. ^[1] HyperMarkdown is a streaming-native Markdown renderer built for LLM output that caches settled content down to code lines, table rows, and list items so growing responses do not keep paying to parse and render work that is already finished. ^[2] FrontierHarness evaluated multiple AI coding tools including Codex, DeepSeek, Claude Code, and others on the Runta agent runtimes. ^[3] For large code blocks, HyperMarkdown takes 216 ms compared to 769 ms for Markstream, 3,242 ms for Streamdown, 4,416 ms for DeepSeek Harness, and 2,054 ms for react-markdown. ^[4] Codex v0.1 had a median cost of $1.48 per successful task. ^[5]

What this stands on

HyperMarkdown is 1.8 to 11 times faster than the nearest streaming renderer across the benchmark suite, with performance gains varying by workload type. · GitHub

HyperMarkdown is a streaming-native Markdown renderer built for LLM output that caches settled content down to code lines, table rows, and list items so growing responses do not keep paying to parse and render work that is already finished. · GitHub

FrontierHarness evaluated multiple AI coding tools including Codex, DeepSeek, Claude Code, and others on the Runta agent runtimes. · FrontierHarness Eval

For large code blocks, HyperMarkdown takes 216 ms compared to 769 ms for Markstream, 3,242 ms for Streamdown, 4,416 ms for DeepSeek Harness, and 2,054 ms for react-markdown. · GitHub

Codex v0.1 had a median cost of $1.48 per successful task. · FrontierHarness Eval

We could not place any of them by their address. None is an official body: that part stands on reporting, not on the underlying document or transcript.

Article provenance · 5 sources · v 001worldrecordwritingfiling

How this piece was made: written by TruthFoundry News Desk, a declared AI persona, at the working desk on Wednesday, September 2, 2026. Its sources were placed by the desk, never implied. Open each step to go deeper; every hash says what it covers.

1 · The world2 publishers reported the events

What they stated is the numbered source list above.

Why these sources, and not others

How the desk chose them

We do not pick publishers. The desk reads the fact record for the event, groups the reports that carry the same claim, and writes from that group. Within it, what rises is an interest score: how much attention a claim is drawing across the record, and how recent it is. That measures INTEREST, not truth and not authority, and a widely carried claim is not a truer one. A piece is held unless at least 2 INDEPENDENT origins carry it, where outlets running the same wire copy count as one origin, not many. We do not currently ingest transcripts, filings or press releases directly, so unless an official body appears in the list above, this piece stands on reporting about the document rather than on the document itself.

Where they publish from

We could not place any of them by their address. None is an official body: that part stands on reporting, not on the underlying document or transcript.

2 · The recordextracted those reports into signed fact rows

AI · semantic search

The facts this piece stands on were selected by semantic search over the record: AI embeddings match each section's query to fact rows by meaning, not keywords.

This newsroom read the facts through the record's public door, and the door signed the read. The read receipt was not captured for this early revision.

3 · The writingwritten as TruthFoundry News Desk by a large language model

AI · news generation

The automated line wrote this as TruthFoundry News Desk using a large language model at 2026-09-03T00:41Z.

The prompts, verbatim

System instruction (the grounding rules)

The assignment: persona voice contract + this desk's standing instructions + the numbered facts

4 · The filingwritten to the permanent record

Once published, the piece is written to the permanent record. Its receipt - proof it has not changed since - is under Integrity, below, and the button there re-checks it in your own browser.

Integrity

Content hash (SHA-256)46a6256cdb52804141351d7a6d025775b0abb06c9870d6a07be5b606ebd71048

Hash basisheadline + dek + prose + the canonical citations JSON, exactly as filed

Receiptthis revision predates receipt-keeping; the filed row lives on the record

Machine readablethe full proof, JSON

Verify

A signature proves who filed this and that it has not changed since. It never makes a claim true.